Optimized design of switching power supply with 89C51 microcontroller as the control core

Introduction

　　The switching power supply is a new type of regulated power supply that uses modern power electronics technology to control the turn-on and turn-off time ratio of the power switch tube (MOSFET, IGBT) to stabilize the output voltage. Since the 1990s, switching power supplies have entered various electronic and electrical equipment fields. Computers, program-controlled switches, communications, electronic testing equipment power supplies, control equipment power supplies, etc. have widely used switching power supplies. Using a switching power supply controlled by a microcontroller can make the switching power supply have more complete functions, further improve intelligence, and facilitate real-time monitoring. Its functions mainly include detecting the running switching power supply and automatically displaying the power status; it can be programmed and controlled through buttons; it can perform fault self-diagnosis and automatically monitor the power part of the power supply; it can protect the power supply from overvoltage and overcurrent; Battery charging and discharging can be controlled in real time.

　　System structure of switching power supply

　　The structure diagram of -48V switching power supply for communication is shown in Figure 1:

　　　　　　　　Figure 1 Structure diagram of switching power supply.

　　After rectification, filtering and power factor correction, the mains power obtains high-voltage direct current, and then the required DC power is obtained through the DC/DC conversion circuit. DC voltage. The control loop samples from the output terminal and compares it with the set reference, then controls the inverter and changes the conduction frequency or conduction/off time of the power switch tube to stabilize the output; on the other hand, according to the data provided by the detection circuit, After identification of the protection circuit, the control circuit is used to perform various protections on the entire machine and charge and discharge control of the battery. The control circuit is the core part of the entire switching power supply. Generally, the control circuit of the switching power supply mainly includes a detection comparison amplifier circuit, a voltage-pulse width conversion circuit (or a voltage-frequency conversion circuit), a clock oscillator (or a constant pulse width generator), It is composed of base drive circuit, overvoltage and overcurrent protection circuit, auxiliary power supply and other circuits. There are shortcomings such as complex circuit, high power consumption, poor sensitivity, and inability to achieve good control.

　　The control circuit composed of the single-chip microcomputer 89C51 module has many advantages such as programmability, strong functions, simple control, and high integration. It also improves the shortcomings of the original circuit. Its principle block diagram is shown in Figure 2.

　　　　　　　　　　　　　　Figure 2 Microcontroller control power supply structure diagram

　　This intelligent switching power supply uses the basic circuit of the switching power supply for communication, with the high-performance microcontroller 89C51 as the control core to form a data processing circuit. With the support of detection and control software, the switching power supply outputs current and voltage. Compare the data sampling with the given data to adjust and control the working status of the switching power tube, while monitoring the output current and performing current control. The working principle of the circuit is: the mains power is converted into direct current by the rectification, filtering and power correction circuit PFC (Power Factor Correct) and sent to the power conversion circuit (DC/DC). The power conversion circuit is between the pulse width modulation circuit (PWM) and the microcontroller. Output stable DC voltage under control. The user can set the voltage value and maximum output current value of the switching power supply through the keyboard as needed. The microcontroller system automatically samples the output voltage and current of the power supply, compares it with the data given by the user, and then controls the switch according to the set adjustment algorithm. Adjust the circuit so that the power supply output voltage meets the given value. While adjusting the output voltage of the power supply, the microcontroller also detects the output current of the circuit. When the output current exceeds a given value, the protection circuit is activated to realize the protection function. In order to make the intelligent switching power supply work reliably and safely, this system is equipped with multiple monitoring and protection systems, mainly including over-current protection and short-circuit protection. The single-chip computer system detects the output current of the switching power tube through a current sensor. When the current exceeds a given value, the single-chip computer system cuts off the switch excitation signal and issues an audible and visual alarm, and detects the battery working condition.

　　Control circuit:

　　The control circuit adopts ATMEL's 89C51 microcontroller, which expands A/D, D/A, keyboard display, and RS232 communication port circuits. The principle structure is shown in Figure 3.

　　　　　　　　　　　　　Figure 3 Control circuit principle structure diagram

　　The control system controls the on and off time of the power conversion switch through D/A conversion through the I/O input port to complete the stabilization of the output voltage, and completes the switching power supply output through A/D conversion. The sampling of voltage and current realizes overvoltage, overcurrent protection and current limiting functions through system software. At the same time, a double closed-loop control system is adopted. When the switching power supply is working, voltage feedback is used to control the output voltage by PWM control to stabilize the output voltage. The control closed loop is a voltage loop or a current loop; when the battery is charging or overloaded, the current signal is used as feedback to control the battery. charge and discharge current and realize overload protection function. In order to accurately control the voltage output of the switching circuit, the high-frequency pulse signal of the microcontroller is divided into a suitable switching pulse signal, which is used as the counting pulse and gate control signal of 89C51. The microcontroller compares the given value with the signal collected by the sensor to generate an error signal. According to the voltage control algorithm setting, 89C51 generates square wave signals with different duty cycles (0~90%), and the circuit voltage output is adjusted through the photocoupler control switch. The output terminal is photoelectrically isolated from the switching circuit, thereby avoiding the impact of disturbance signals from the switching power supply circuit on the normal operation of the microcontroller system.

　　In view of the high precision and fast adjustment characteristics of the controlled switching circuit output voltage, an improved PID control algorithm can be used, which has the advantages of fast voltage adjustment, small overshoot, and stable performance. The keyboard and display part are installed on the instrument operation panel and consist of 8-digit LED digital tubes, 3 LED indicator lights and 16 keys. 4-digit digital tubes display the power supply voltage, 4-digit digital tubes display current, and 3 LED indicator lights. Displayed as an alarm.

　　System software design

　　This software mainly completes signal sampling, various data processing, and control of the power conversion part. The system software mainly includes key switch scanning program, fault identification subroutine, equalizing charging and floating charging subroutine, interruption detection subroutine and communication subroutine, etc. The main program flow chart is shown in Figure 4.

　　　　　　　　　　　Figure 4 Main program flow chart

　　During the initialization process, each input port of the 89C51 is first reset, and then the data stored before the last shutdown is read from the EEROM, the switching circuit is controlled, and displayed. After the initialization is completed, open the interrupt program. If there is an interrupt request, respond, otherwise perform data sampling and read the given value, and then perform data processing; if there is a short circuit or overcurrent situation, call the alarm protection subroutine; if you want to maintain a certain dynamics of the battery float, you can To a certain extent, it reflects the internal changes in the battery and the size of the SoC. However, this method assumes that the current is time-varying during the derivation process. If the battery is discharged at a constant current for a long period of time, it will greatly reduce the prediction accuracy of the SoC. accuracy. The dynamic model based on the state space builds a model based on the dynamic changes of the reactants, uses the measured current and voltage as input quantities to calculate the SoC, and also considers the diffusion phenomenon of active substances to improve the accuracy of the SoC. This is a better method ; However, due to the high order of the battery model, the calculation is difficult. The establishment of the model requires the determination of a considerable number of empirical parameters, which brings great trouble to the application.

　　The SoC definition based on the energy model corrects the shortcomings of the original SoC model. Taking into account the recoverability of the battery, it integrates current, voltage, and resistance judgments, which improves the judgment accuracy of SoC to a certain extent, but it does not consider the influence of temperature. A large amount of experimental data is needed. Since the battery is sealed, the only external measurable parameters are current and voltage. The Randels Ershler battery model is used to model the battery, and the SoC is estimated through accurate ampere-hour integration, while capacity aging compensation, temperature compensation, self-discharge compensation and discharge are performed. Rate compensation is also a feasible method.

　　The above methods can reflect the remaining power to a certain extent and are suitable for prediction of battery SoC for electric vehicles. However, determining these model parameters requires many iterative steps, and importantly, these algorithms must know the initial value of the battery's SoC. Because it takes time to calculate and display the SoC value in real time. The more complex the model, the more time it takes to calculate the SoC. There are many SoC prediction methods, but to achieve higher accuracy, there is still a lot of work to be done in battery modeling and SoC prediction methods.